基于校验函数梯度的Turbo码编码器识别  

Turbo code encoder identification based on check function gradient

作　　者：陈增茂 刘美祯 孙志国[1] 孙溶辰 CHEN Zengmao;LIU Meizhen;SUN Zhiguo;SUN Rongchen(College of Information and Communication Engineering,Harbin Engineering University,Harbin 150001,China;Key Laboratory of Advanced Marine Communication and Information Technology(Harbin Engineering University),Ministry of Industry and Information Technology,Harbin 150001,China)

机构地区：[1]哈尔滨工程大学信息与通信工程学院,黑龙江哈尔滨150001 [2]哈尔滨工程大学工业和信息化部先进船舶通信与信息技术重点实验室,黑龙江哈尔滨150001

出　　处：《哈尔滨工程大学学报》2024年第9期1826-1833,共8页Journal of Harbin Engineering University

基　　金：国家自然科学基金项目(62001139).

摘　　要：针对现有的Turbo码编码器识别算法中存在容错性差和计算量大的问题,本文提出了计算复杂度低和在低信噪比条件下容错性强的编码器识别算法。本文利用码元之间线性约束关系,构造代价函数,定义每个时刻的函数符合度,这个概念能明确地展现每个时刻码元之间约束关系的强弱;函数符合度将作为识别递归系统卷积码的代价函数,待识别的生成多项式参数的概率值作为代价函数的变量,将RSC码识别问题进而转化为代价函数求解极大值问题。利用迭代更新的方式,在迭代次数一定的条件下,对函数中的未知生成矩阵参数求解梯度,完成对未知生成矩阵参数的识别。仿真结果表明:算法在迭代20次时可以对编码器参数的识别收敛,同时在低信噪比条件下相比已有算法容错性强、计算复杂度低。To overcome difficulties in poor performances of fault tolerance and high computational effort in existing turbo code encoder recognition algorithms,an encoder recognition algorithm with low computational complexity and high fault tolerance under low signal-to-noise ratio(SNR)conditions is proposed.In this paper,the linear constraint relationship among symbols is used to construct a coding cost function,and the function conformity at each moment is defined.This concept clearly shows the strength of the constraint relationship among symbols at each moment.The function conformity is used as the cost function to identify the recursive systematic convolutional(RSC)code,the probability of the generated polynomial parameter to be discerned is regarded as the variable of the cost function,and the RSC code identification problem is then transformed into solving a maximal value problem of the cost function.Finally,using iterative updating,under a fixed amount of iterations,the gradient of the un-known generator matrix parameters in the function is solved to complete the identification.The simulation results demonstrate that the proposed algorithm can recognize encoder parameters when iterating 20 times.Moreover,the computational complexity is low,and the error tolerance is stronger than those of existing algorithms in low-SNR scenarios.

关 键 词：低信噪比 函数符合度 递归系统卷积码 梯度 

分 类 号：TN911.7[电子电信—通信与信息系统]